most recent 30 from http://mathoverflow.net 2013-05-21T18:28:50Z http://mathoverflow.net/feeds/user/17225 http://www.creativecommons.org/licenses/by-nc/2.5/rdf http://mathoverflow.net/questions/73135/degrees-of-freedom-for-chi-squared-test Susanna 2011-08-18T11:15:34Z 2011-08-18T11:15:34Z

<p>Hi,</p> <p>I am wonder about the usage of degrees of freedom for the chi-squared distribution. </p> <p>I am interested in comparing observations to expected values, thus limiting my focus on the Single Sample Chi Square Test. My concern is that I miscalculate the d.f. which is usually defined as the the number of terms for $\chi^2$. </p> <p>Here is an extract:</p> <p>(observed data)</p> <p>(period) PlayerI PlayerII PLayerIII</p> <p>1900-1950 50 60 9</p> <p>1901-2000 60 29 23</p> <p>(expected data)</p> <p>(period) PlayerI PlayerII PLayerIII</p> <p>1900-1950 20.4 5.2 24.5</p> <p>1901-2000 99.4 24.5 8.2</p> <p>If i were to use the d.f. of freedom , would I deal with d.f.=6-1=5 (because in the calculation for $\chi^2$, I will wind up with 6 terms right. </p> <p>or is d.f.=3-1=1 (based on the players).</p>